Am 31.01.2017 um 22:51 schrieb Gianluca Cecchi:
> Hello,
> suppose I want to use a special configuration for my IBM/1814 storage array
> luns, then I put something like this in multipath.conf
>
> devices {
> device {
> vendor "IBM"
> product "^1814"
> product_blacklist "Universal Xport"
> path_grouping_policy "group_by_prio"
> path_checker "rdac"
> features "0"
> hardware_handler "1 rdac"
> prio "rdac"
> failback immediate
> rr_weight "uniform"
> no_path_retry "12"
> }
> }
>
> In CentOS 6.x when you restart multipathd or restart server, using
>
> multipathd -k
> multipathd> show config
> multipathd> exit
That prints out the build-in defaults. See
https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/7/pdf/DM_Multipath/Red_Hat_Enterprise_Linux-7-DM_Multipath-en-US.pdf
chapter 4.
> in output you see your particular configuration instead than the default
> one for this particular device.
> In CentOS 7.3 instead I see it two times, the first one with the default
> built-in values (eg "no_path_retry fail") and at the end the customized
> one.
> So it is not clear what is the actual configuration that device mapper
> multipath is using...
> The last wins?
> Is this expected behaviour? In that case what command can I use to
> crosscheck it (apart from real testing that is anyway necessary to verify
> too)?
> Eg in CentOS 7.3 I'm using device-mapper-multipath-0.4.9-99.el7_3.1.x86_64
> while on CentOS 6.8 I'm using device-mapper-multipath-0.4.9-93.el6.x86_64
>
> Thanks in advance,
> Gianluca
To configure a custom setup and enable it please see chapter 3 of the
RHEL documentation about device-mapper multipath.
Regards
Alexander